Adaptive Knowledge Platform

Abstract

Methods, systems, and apparatus, including medium-encoded computer program products, for providing an adaptive knowledge platform. In one or more aspects, a system can include a knowledge management component to acquire, classify and disseminate information of a dataset; a human-computer interaction component to visualize multiple perspectives of the dataset and to model user interactions with the multiple perspectives; and an adaptivity component to modify one or more of the multiple perspectives of the dataset based on a user-interaction model.

Description

BACKGROUND[0001]This specification relates to organizing, filtering, and accessing content in a content management system.[0002]Various knowledge management systems have been implemented using different approaches to content classification, as well as different approaches to viewing the data contained in the system. Numerous methods have been developed to address content categorization and visualization. These methods have included the use of both symbolic knowledge representation and statistical machine learning techniques.[0003]A symbolic knowledge representation is typically referred to as an ontology. In computer science, an ontology generally refers to a hierarchical knowledge structure that contains a vocabulary of terms and concepts for a specific knowledge domain and contains relevant interrelationships between those terms and concepts, typically represented in a tree data structure. A traditional symbolic knowledge ontology is typically constructed by hand by one or more do...

AI Technical Summary

Benefits of technology

[0012]Particular embodiments of the subject matter described in this specification can be implemented to realize one or more of the following advantages. Content, which can be stored in computer systems and networks, can be readily organized, filtered, and accessing in virtually any knowledge domain. A content management system can bring together knowledge management, adaptivity, and human-computer interaction in a deeply integrated way. Multiple systems and subsystems can be integrated into a content neutral knowledge platform, which can be used to organize, filter and access a dataset that pertains to multiple knowledge domains.

[0013]Core systems and subsystems can support powerful visualization, perspective-driven data organization, dynamic content growth, search, adaptivity, and evaluation capabilities. The knowledge platform can provide one or more of the following capabilities. Large bodies of content can be visualized intuitively and clearly, with visual representations suggestive of meaning. Navigation through the content can be facilitated and informed by the system so users will less likely feel lost. Contextual information can be provided to enhance understanding and potential knowledge discovery. Semantic search of a given dataset, and metasearch of sources outside the given dataset, can be supported. Users can designate multilinear pathways through the content of the dataset. User-created pathways designating instructional approaches through content can be supported. Multiple perspectives of the dataset can be displayed, as well as intersecting disciplines and points of view within the dataset. Quantitative content can be integrated with qualitative perspectives. A system for hypothesis creation and testing can be realized. The knowledge platform can efficiently evolve with new content—both automatically and through user-collaboration. Content delivery can be automatically custom-tailored to a range of user needs. Moreover, user response and effectiveness of the content on the user can be evaluated.

[0014]A knowledge platform can integrate dynamic content organization (such as automatic content acquisition, classifying, and filtering), contextual browsing, and customized content delivery. It can allow users to see the whole picture of content (and see both the proverbial forest of information as well as the individual trees); spot gaps in data sets and relationships; easily access the content they need when they need it; visualize critical data factors that are important to users (such as, for instance, the dimension of time); and provide facilitated and informed navigation through data fields. Accessing and utilizing information can be facilitated with tools that support: the accumulation of new, relevant knowledge classified to facilitate maximum understanding and accessibility; the archiving and dissemination of content within relational contexts of meaning; the personalization of content delivery; and the ability to integrate quantitative and qualitative information. Content provided by the system can be made easier to access, more efficient to search or browse, and simpler for users to identify what they need when they need it. A broad contextual perspective on information can be provided, and fewer limitations need be placed on how content is visualized in relation to other content. A wide variety of user needs for personalized information delivery can be addressed and responded to, the display of multiple perspectives of a given body of content can be supported, multiple sets of quantitative data can be readily and intuitively integrated with qualitative information, hypothesis creation, modeling and testing can be facilitated, and user response to, and the effectiveness of, content delivered within a nonlinear hierarchy context can be evaluated. The knowledge platform can employ annotation tools that support both key word and semantic annotation in an integrated easy-to-use system. An automatic knowledge acquisition system can be bundled with other technologies in a way that supports semantic annotation and personalized information classification, filtering, and delivery.

[0015]A knowledge platform, as described herein, can be implemented as a content management system for multiple types of datasets and within various application contexts. For example, the systems and techniques described herein can be implemented in a multimedia archive management system used to archive and retrieve multimedia content. This can provide an integrated approach to present content in a given dataset that frames the parts in context of the whole, even as it provides for easy access, filtering, and dissemination.

Problems solved by technology

This approach is scalable but can be very sensitive to the data in the document training set.

Images